@inproceedings{glocker-markianos-wright-2020-textmarkers,
title = {{T}{\"e}{X}tmarkers at {S}em{E}val-2020 Task 10: Emphasis Selection with Agreement Dependent Crowd Layers},
author = "Glocker, Kevin and
Markianos Wright, Stefanos Andreas",
booktitle = "Proceedings of the Fourteenth Workshop on Semantic Evaluation",
month = dec,
year = "2020",
address = "Barcelona (online)",
publisher = "International Committee for Computational Linguistics",
url = "https://aclanthology.org/2020.semeval-1.222",
doi = "10.18653/v1/2020.semeval-1.222",
pages = "1698--1703",
abstract = "In visual communication, the ability of a short piece of text to catch someone{'}s eye in a single glance or from a distance is of paramount importance. In our approach to the SemEval-2020 task {``}Emphasis Selection For Written Text in Visual Media{''}, we use contextualized word representations from a pretrained model of the state-of-the-art BERT architecture together with a stacked bidirectional GRU network to predict token-level emphasis probabilities. For tackling low inter-annotator agreement in the dataset, we attempt to model multiple annotators jointly by introducing initialization with agreement dependent noise to a crowd layer architecture. We found our approach to both perform substantially better than initialization with identities for this purpose and to outperform a baseline trained with token level majority voting. Our submission system reaches substantially higher Match m on the development set than the task baseline (0.779), but only slightly outperforms the test set baseline (0.754) using a three model ensemble.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="glocker-markianos-wright-2020-textmarkers">
<titleInfo>
<title>TëXtmarkers at SemEval-2020 Task 10: Emphasis Selection with Agreement Dependent Crowd Layers</title>
</titleInfo>
<name type="personal">
<namePart type="given">Kevin</namePart>
<namePart type="family">Glocker</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Stefanos</namePart>
<namePart type="given">Andreas</namePart>
<namePart type="family">Markianos Wright</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2020-dec</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Fourteenth Workshop on Semantic Evaluation</title>
</titleInfo>
<originInfo>
<publisher>International Committee for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Barcelona (online)</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>In visual communication, the ability of a short piece of text to catch someone’s eye in a single glance or from a distance is of paramount importance. In our approach to the SemEval-2020 task “Emphasis Selection For Written Text in Visual Media”, we use contextualized word representations from a pretrained model of the state-of-the-art BERT architecture together with a stacked bidirectional GRU network to predict token-level emphasis probabilities. For tackling low inter-annotator agreement in the dataset, we attempt to model multiple annotators jointly by introducing initialization with agreement dependent noise to a crowd layer architecture. We found our approach to both perform substantially better than initialization with identities for this purpose and to outperform a baseline trained with token level majority voting. Our submission system reaches substantially higher Match m on the development set than the task baseline (0.779), but only slightly outperforms the test set baseline (0.754) using a three model ensemble.</abstract>
<identifier type="citekey">glocker-markianos-wright-2020-textmarkers</identifier>
<identifier type="doi">10.18653/v1/2020.semeval-1.222</identifier>
<location>
<url>https://aclanthology.org/2020.semeval-1.222</url>
</location>
<part>
<date>2020-dec</date>
<extent unit="page">
<start>1698</start>
<end>1703</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T TëXtmarkers at SemEval-2020 Task 10: Emphasis Selection with Agreement Dependent Crowd Layers
%A Glocker, Kevin
%A Markianos Wright, Stefanos Andreas
%S Proceedings of the Fourteenth Workshop on Semantic Evaluation
%D 2020
%8 dec
%I International Committee for Computational Linguistics
%C Barcelona (online)
%F glocker-markianos-wright-2020-textmarkers
%X In visual communication, the ability of a short piece of text to catch someone’s eye in a single glance or from a distance is of paramount importance. In our approach to the SemEval-2020 task “Emphasis Selection For Written Text in Visual Media”, we use contextualized word representations from a pretrained model of the state-of-the-art BERT architecture together with a stacked bidirectional GRU network to predict token-level emphasis probabilities. For tackling low inter-annotator agreement in the dataset, we attempt to model multiple annotators jointly by introducing initialization with agreement dependent noise to a crowd layer architecture. We found our approach to both perform substantially better than initialization with identities for this purpose and to outperform a baseline trained with token level majority voting. Our submission system reaches substantially higher Match m on the development set than the task baseline (0.779), but only slightly outperforms the test set baseline (0.754) using a three model ensemble.
%R 10.18653/v1/2020.semeval-1.222
%U https://aclanthology.org/2020.semeval-1.222
%U https://doi.org/10.18653/v1/2020.semeval-1.222
%P 1698-1703
Markdown (Informal)
[TëXtmarkers at SemEval-2020 Task 10: Emphasis Selection with Agreement Dependent Crowd Layers](https://aclanthology.org/2020.semeval-1.222) (Glocker & Markianos Wright, SemEval 2020)
ACL